Core dump when using smoothxg from conda::pggb-0.2.0-hdfd78af_0

[songtaogui]

Hi, Erik

Thank you for providing lots of wonderful pangenome related tools.

Recently, I was trying the pggb pipeline. However, the pipeline failed in the smoothxg step, with Illegal instruction (core dumped) error, after checking the generated core. file, I got the message below:

Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `smoothxg -t 8 -T 8 -g pggb/all.fa.57209dc.2ff309f.seqwish.gfa -w 39351 -K -X 10'.
Program terminated with signal SIGILL, Illegal instruction.
#0  0x000055643e10c127 in std::__fill_bvector (__x=<optimized out>, __last=13, __first=0, __v=0x2b930e9cf788)
    at /opt/conda/conda-bld/smoothxg_1637224140228/_build_env/x86_64-conda-linux-gnu/include/c++/9.4.0/bits/stl_bvector.h:401
401     /opt/conda/conda-bld/smoothxg_1637224140228/_build_env/x86_64-conda-linux-gnu/include/c++/9.4.0/bits/stl_bvector.h: No such file or directory.

My conda environment was created with pggb-0.2.0-hdfd78af_0.

I have also ran the same pipeline using the docker image, and got the same core dump error.

So, what did I miss and how could I fix the error?

Please also find attached the log files for the pggb pipe if it would help. pggb.log

Thank you

Best wishes

Songtao Gui

[ekg]

This looks like a binary incompatibility with your processor.

Is there any chance you could run the docker build locally? That'd fix this.

On Sun, Jan 16, 2022, 19:21 Songtao Gui @.***> wrote:

Hi, Erik

Thank you for providing lots of wonderful pangenome related tools.

Recently, I was trying the pggb pipeline. However, the pipeline failed in the smoothxg step, with Illegal instruction (core dumped) error, after checking the generated core. file, I got the message below:

Using host libthread_db library "/lib64/libthread_db.so.1". Core was generated by `smoothxg -t 8 -T 8 -g pggb/all.fa.57209dc.2ff309f.seqwish.gfa -w 39351 -K -X 10'. Program terminated with signal SIGILL, Illegal instruction.

0 0x000055643e10c127 in std::__fill_bvector (x=, last=13, first=0, v=0x2b930e9cf788)

at /opt/conda/conda-bld/smoothxg_1637224140228/_build_env/x86_64-conda-linux-gnu/include/c++/9.4.0/bits/stl_bvector.h:401

401 /opt/conda/conda-bld/smoothxg_1637224140228/_build_env/x86_64-conda-linux-gnu/include/c++/9.4.0/bits/stl_bvector.h: No such file or directory.

My conda environment was created with pggb-0.2.0-hdfd78af_0.

I have also ran the same pipeline using the docker image, and got the same core dump error.

So, what did I miss and how could I fix the error?

Please also find attached the log files for the pggb pipe if it would help. pggb.log https://github.com/pangenome/pggb/files/7877620/pggb.log

Thank you

Best wishes

Songtao Gui

— Reply to this email directly, view it on GitHub https://github.com/pangenome/pggb/issues/152, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABDQENGZH3YKVCGL6PCML3UWMEDHANCNFSM5MDBSRIQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you are subscribed to this thread.Message ID: @.***>

[songtaogui]

:Thank you for your kind help.

Unfortunately, I cannot run the docker build locally due to the root permissions on out cluster. Instead, I build a singularity local image from the pggb docker hub and ran it locally. However, I again encountered the same error:

[ekg]

Can you build the individual tools on your system? Can your administrator build them for you?

There is a processor level incompatibility. In general this is only possible to resolve if you can build on the system itself.

On Mon, Jan 17, 2022, 03:39 Songtao Gui @.***> wrote:

:Thank you for your kind help.

Unfortunately, I cannot run the docker build locally due to the root permissions on out cluster. Instead, I build a singularity local image from the pggb docker hub and ran it locally. However, I again encountered the same error: [image: pggb-sglrty] https://user-images.githubusercontent.com/14274330/149699720-e644d5a0-11f1-4ab7-9a8b-99a13e06c386.png

— Reply to this email directly, view it on GitHub https://github.com/pangenome/pggb/issues/152#issuecomment-1014085682, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABDQEKYWH7ZTLQBMXDMGXLUWN6OPANCNFSM5MDBSRIQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you commented.Message ID: @.***>

[AndreaGuarracino]

@songtaogui, do you still have the same problems with the last pggb 0.3.1 on Docker/bioconda?

[AndreaGuarracino]

Version 0.3.1 on bioconda fixed a number of compatibility issues.